FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2041316200> ?p ?o ?g. }

• W2041316200 endingPage "13665" @default.
• W2041316200 startingPage "13660" @default.
• W2041316200 abstract "Membrane vesicles were prepared fromEscherichia coli cells in which SecE was depleted to 2% of wild-type membranes. SecE depletion had pleiotropic effects; SecD, SecF, SecG, and SecY were decreased 4–6-fold, whereas SecA was increased about 16-fold over that of wild-type membranes. These membranes were substantially active in the in vitrotranslocation of proOmpA, which was mediated by the SecA pathway since it was inhibited by azide. Similar substantial translocation activities were observed for proLamB and proLpp in the SecE-depleted membranes. However, the translocation of proPhoA was more severely impaired. These data indicate that SecE may enhance but is not obligatorily required for the translocation of at least certain precursors, and suggest that the effects of the SecE depletion on protein translocation may be precursor-dependent. Membrane vesicles were prepared fromEscherichia coli cells in which SecE was depleted to 2% of wild-type membranes. SecE depletion had pleiotropic effects; SecD, SecF, SecG, and SecY were decreased 4–6-fold, whereas SecA was increased about 16-fold over that of wild-type membranes. These membranes were substantially active in the in vitrotranslocation of proOmpA, which was mediated by the SecA pathway since it was inhibited by azide. Similar substantial translocation activities were observed for proLamB and proLpp in the SecE-depleted membranes. However, the translocation of proPhoA was more severely impaired. These data indicate that SecE may enhance but is not obligatorily required for the translocation of at least certain precursors, and suggest that the effects of the SecE depletion on protein translocation may be precursor-dependent. The general secretion pathway of SecA-dependent protein translocation in Escherichia coli has been studied extensively (1Pugsley A.P. Microbiol. Rev. 1993; 57: 50-108Google Scholar). The Sec machinery is generally believed to be composed of several inner membrane proteins: SecA, SecD, SecE, SecF, SecG, and SecY (PrlA). SecA plays a central role in this machinery, and its requirement for secretion has been well established (2Cabelli R.J. Chen L. Tai P.C. Oliver D.B. Cell. 1988; 55: 683-692Google Scholar, 3Lill R. Cunningham K. Brundage L.A. Ito K. Oliver D. Wickner W. EMBO J. 1989; 8: 961-966Google Scholar, 4Watanabe M. Blobel G. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 9011-9015Google Scholar). SecA initiates an early step of translocation by hydrolyzing ATP (5Schiebel E. Driessen A.J.M. Hartl F. Wickner W. Cell. 1991; 64: 927-939Google Scholar), after forming a complex with secretory precursors and the cytoplasmic chaperone, SecB (6Hartl F. Lecker S. Schiebel E. Hendrick J.P. Wickner W. Cell. 1990; 63: 269-279Google Scholar). The energy supplied by ATP hydrolysis is believed to drive the insertion and deinsertion of SecA onto and off the plasma membrane (7Economou A. Wickner W. Cell. 1994; 78: 835-843Google Scholar, 8Economou A. Pogliano J.A. Beckwith J. Oliver D.B. Wickner W. Cell. 1995; 83: 1171-1181Google Scholar). However, recent data indicate that some SecAs are permanently integrated into membrane and are active in protein translocation (9Kim Y.J. Rajapandi T. Oliver D. Cell. 1994; 78: 845-853Google Scholar, 10Chen X. Xu H. Tai P.C. J. Biol. Chem. 1996; 271: 29698-29706Google Scholar). SecD and SecF are also integral membrane proteins and are necessary for efficient protein translocation but not for cell viability (11Pogliano J.A. Beckwith J. EMBO J. 1994; 13: 554-561Google Scholar). Both are presumed to be involved in a late step of the translocation reaction (12Matsuyama S. Fujita Y. Mizushima S. EMBO J. 1993; 12: 265-270Google Scholar, 13Arkowitz R.A. Wickner W. EMBO J. 1994; 13: 954-963Google Scholar). However, they do not exhibit significant effects on the translocation activity of reconstituted membranes (14Matsuyama S. Fujita Y. Sagara K. Mizushima S. Biochim. Biophys. Acta. 1992; 1122: 77-84Google Scholar). Together, SecE, SecG, and SecY appear to form an integral part of the translocation machinery in the inner membrane since these proteins are copurified chromatographically and coimmunoprecipitated with anti-SecY antibody (15Brundage L. Hendrick J.P. Schiebel E. Driessen A.J.M. Wickner W. Cell. 1990; 62: 649-657Google Scholar, 16Brundage L. Fimmel C.J. Mizushima S. Wickner W. J. Biol. Chem. 1992; 267: 4166-4170Google Scholar, 17Nishiyama K. Mizushima S. Tokuda H. EMBO J. 1993; 12: 3409-3415Google Scholar). Additionally, when overexpressed, these three proteins can be recovered in the plasma membrane fraction (18Douville K. Price A. Eichler J. Economou A. Wickner W. J. Biol. Chem. 1995; 270: 20106-20111Google Scholar). However, SecG is not essential for translocation since reconstituted membranes without SecG are still active in protein export, although the translocation activity can be enhanced by the addition of SecG (19Hanada M. Nishiyama K. Mizushima S. Tokuda H. J. Biol. Chem. 1994; 269: 23625-23631Google Scholar). In addition, strains deleted of secG grow normally but are cold-sensitive (20Nishiyama K. Hanada M. Tokuda H. EMBO J. 1994; 13: 3272-3277Google Scholar). On the other hand, SecE and SecY are thought to be indispensable components of protein export in reconstituted membranes (15Brundage L. Hendrick J.P. Schiebel E. Driessen A.J.M. Wickner W. Cell. 1990; 62: 649-657Google Scholar, 21Akimaru J. Matsuyama S. Tokuda H. Mizushima S. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 6545-6549Google Scholar, 22Nishiyama K. Kabuyama Y. Akimaru J. Matsuyama S. Tokuda H. Mizushima S. Biochim. Biophys. Acta. 1991; 1065: 89-97Google Scholar). Nevertheless, it has been reported that reconstituted membranes without SecY are active in protein translocation (4Watanabe M. Blobel G. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 9011-9015Google Scholar, 23Watanabe M. Nicchitta C.V. Blobel G. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 1960-1964Google Scholar). The requirement for SecE is addressed in this study. SecE is a 13.6-kDa integral membrane protein, spanning E. coli inner membrane three times with its amino terminus in the cytoplasm (24Schatz P.J. Riggs P.D. Jacq A. Fath M.J. Beckwith J. Genes Dev. 1989; 3: 1035-1044Google Scholar). Complementation studies revealed that the third transmembrane segment and its adjacent cytoplasmic domain is sufficient for the function of SecE (25Schatz P.J. Bieker K.L. Ottemann K.M. Silhavy T.J. Beckwith J. EMBO J. 1991; 10: 1749-1757Google Scholar, 26Murphy C.K. Beckwith J. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 2557-2561Google Scholar). Genetic data showed that the secE gene product is essential for cell growth (25Schatz P.J. Bieker K.L. Ottemann K.M. Silhavy T.J. Beckwith J. EMBO J. 1991; 10: 1749-1757Google Scholar), and its mutation causes precursor accumulation (27Riggs P.D. Derman A.I. Beckwith J. Genetics. 1988; 118: 571-579Google Scholar). In this study, we depleted cells of SecEin vivo to elucidate the role of SecE in protein translocation in vitro. The results show that membranes prepared from SecE-depleted cells still have substantial translocation activity for several precursors, indicating that SecE is not essential for in vitro translocation of certain secretory proteins. Wild-type E. coliMC1000 (28Oliver D.B. Beckwith J. Cell. 1982; 30: 311-319Google Scholar) membranes were prepared as described (29Tai P.C. Tian G. Xu H. Lian J.P. Yu J.N. Methods Cell Biol. 1991; 34: 167-187Google Scholar). SecE-depleted membranes were prepared from E. coli strain PS289 (MC1000,leu +, ara +,phoAΔPvuII, pcnB80,zadL::Tn10 (Tcs Strr),secEΔ19–111,recA::cat/pBAD22secE +), a gift from C. Murphy and J. Beckwith, Harvard Medical School, Boston, MA (30Traxler B. Murphy C. J. Biol. Chem. 1996; 271: 12394-12400Google Scholar), carrying a chromosomal deletion in secE and a plasmid with secE behind an arabinose-activable promoter. PS289 cells were grown to mid-log phase at 37 °C in LinA medium (29Tai P.C. Tian G. Xu H. Lian J.P. Yu J.N. Methods Cell Biol. 1991; 34: 167-187Google Scholar) containing 0.2%l-arabinose, harvested, washed twice, and resuspended to the same medium without arabinose. Cells were inoculated to 0.1A600 unit in LinA supplemented with either 0.5% glucose or 0.2% arabinose, and harvested after 4 doublings (for arabinose medium) or after growth ceased (for glucose medium). Inverted inner membrane vesicles from PS289 cells were prepared as described (29Tai P.C. Tian G. Xu H. Lian J.P. Yu J.N. Methods Cell Biol. 1991; 34: 167-187Google Scholar). 35S-proLamB and 35S-proPhoA were synthesized from plasmid pEB (lamB under control of pT7–6, constructed in this laboratory) and pAE2.2 (phoAunder control of T7 RNA polymerase promoter, a gift from D. Oliver, Wesleyan University, Middletown, CT), respectively, using an in vitro transcription-translation coupled system (31Zhong X. Kolter R. Tai P.C. J. Biol. Chem. 1996; 271: 28057-28063Google Scholar) with addition of 2% polyethylene glycol 6000. 35S-proOmpA and35S-proLpp were synthesized with mRNA prepared from OmpA- and Lpp-overproducing cells, respectively, as described (29Tai P.C. Tian G. Xu H. Lian J.P. Yu J.N. Methods Cell Biol. 1991; 34: 167-187Google Scholar). After translation, ribosomes and endogenous membranes were removed by centrifugation, and this posttranslational supernatant was used as labeled precursors. A typical 100-μl translocation reaction contained 70 μl of 35S-labeled precursor solution, 2 μl of an ATP-regenerating stock (29Tai P.C. Tian G. Xu H. Lian J.P. Yu J.N. Methods Cell Biol. 1991; 34: 167-187Google Scholar), and 25 μg of inverted inner membrane vesicles. Purified SecA was added to 20 μg/ml where indicated. The reaction was carried out at 37 °C for 10 min unless otherwise indicated, and then followed the procedures as described (29Tai P.C. Tian G. Xu H. Lian J.P. Yu J.N. Methods Cell Biol. 1991; 34: 167-187Google Scholar). Membranes (40 μg) were suspended in 100 μl of 0.1 mNa2CO3, or of 6 m urea, incubated at 0 °C for 30 min, then recovered by centrifugation with Beckman TLA 100.2 rotor at 73,000 rpm for 2 h over a 200-μl cushion (0.5m sucrose in DTK buffer (1 mm dithiothreitol, 10 mm Tris-HCl, pH 7.6, and 50 mm KCl)). Heparin treatment was the same as described (4Watanabe M. Blobel G. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 9011-9015Google Scholar). Purified SecA and 35S-SecA are from laboratory stock purified from pT7-secA (2Cabelli R.J. Chen L. Tai P.C. Oliver D.B. Cell. 1988; 55: 683-692Google Scholar). ProOmpA was purified from BA13/pAM103 (2Cabelli R.J. Chen L. Tai P.C. Oliver D.B. Cell. 1988; 55: 683-692Google Scholar, 10Chen X. Xu H. Tai P.C. J. Biol. Chem. 1996; 271: 29698-29706Google Scholar). Purified SecB was from laboratory stock obtained as described (32Weiss J.B. Ray P.H. Bassford Jr., P.J. Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 8978-8982Google Scholar). SDS-PAGE 1The abbreviations used are: PAGE, polyacrylamide gel electrophoresis; Tricine,N-tris(hydroxymethyl)methylglycine. was carried out according to Laemmli (33Laemmli U.K. Nature. 1970; 227: 680-685Google Scholar). For Lpp, a Tricine-15% SDS-PAGE system was performed by the method of Schagger and von Jagow (34Schagger H. von Jagow G. Anal. Biochem. 1987; 166: 368-379Google Scholar). Protein amounts were determined by Bio-Rad assay kit according to Bradford (35Bradford M.M. Anal. Biochem. 1976; 72: 248-254Google Scholar) with γ-globulin as a standard. All autoradiogram quantitations were carried out by PDI Image Analyzing System (Protein Databases Inc., Huntington Station, NY). Proteinase K was from Boehringer Mannheim. [35S]Methionine was from DuPont NEN. All other chemicals are of reagent grade and were obtained from Sigma or other commercial sources. Without arabinose, growth of PS289 cells ceased after 3 doublings, which is consistent with the findings that secEis essential for cell growth (25Schatz P.J. Bieker K.L. Ottemann K.M. Silhavy T.J. Beckwith J. EMBO J. 1991; 10: 1749-1757Google Scholar). Inverted inner membrane vesicles were prepared from these cells, and all Sec proteins in the membranes were examined by immunoblots. The depletion of SecE in these membranes was confirmed (Fig. 1). Surprisingly, not only SecY and SecG decreased as expected; SecD and SecF were also reduced (Fig. 1). The amount of membrane used for immunoblotting was adjusted to visualize the relative levels of Sec proteins, and were quantitated (Table I). The results revealed that membranes isolated from PS289 cells grown with arabinose (referred to as SecE++ membranes hereafter), contained around 12-fold more SecE (expressed from multicopy plasmid), 2-fold more SecA and SecG, and normal amounts of SecD, SecF, and SecY as compared with isogenic wild-type MC1000 membranes. SecE in the membranes from PS289 cells shifted from arabinose to glucose media was depleted 600-fold and reduced to about 2% of wild-type MC1000 membranes. The depletion of SecE had pleiotropic effects on other Sec proteins: SecD, SecY, SecF, and SecG also decreased to about 25, 15, 22, and 17%, respectively. In contrast, SecA increased about 16-fold in these membranes. This increase was due to increased expression and not to the redistribution of SecA protein between the cytosol and the membranes (data not shown).Table IRelative Sec protein amountsMembranesSecASecDSecESecFSecGSecYMC1000100100100100100100PS289 (+Ara)197 ± 23107 ± 61193 ± 259110 ± 9171 ± 8103 ± 5PS289 (−Ara)1583 ± 24625 ± 22.3 ± 0.422 ± 217.4 ± 0.515 ± 2Experiments were performed as described in the legend to Fig. 1 except that membrane amounts were adjusted to obtain appropriate signals. Data are presented as mean ± S.E. of nine independent experiments. The Sec protein amounts in wild-type MC1000 membranes were referred to 100%. Open table in a new tab Experiments were performed as described in the legend to Fig. 1 except that membrane amounts were adjusted to obtain appropriate signals. Data are presented as mean ± S.E. of nine independent experiments. The Sec protein amounts in wild-type MC1000 membranes were referred to 100%. SecE-depleted membranes were examined for in vitro proOmpA translocation activity. Surprisingly, SecE-depleted membranes had the same translocation activity as wild-type MC1000 membranes with either SecA-depleted S30 or normal SecA-containing S30 that were used to synthesize nascent proOmpA (Fig. 2). More importantly, translocation in SecE-depleted membranes included the processing of precursor form to mature OmpA, and the translocated proteins were still resistant to 1 mg/ml proteinase K, indicating that SecE-depleted membranes were not porous and behaved similarly to normal membranes in translocation. (The SecE-depleted membranes, however, were more sensitive to freeze-thawing.) Since SecE-depleted membranes had 16 times more SecA than wild-type membranes, this might account for the high activity of these membranes. To test for this, additional SecA was added in translocation assays. The translocation activity of wild-type MC1000 or SecE++ membranes increased up to 250%, reaching saturation at 20 μg/ml additional SecA (Fig.3 A). For SecE-depleted membranes, 20 μg/ml SecA only enhanced translocation activity to a maximum of 1.4-fold. Even under this condition, the SecE-depleted membranes still had about 50% translocation activity of MC1000 membranes. The lower enhancement of translocation activity for SecE-depleted membranes by SecA was not due to reduced binding of SecA, since SecE-depleted membranes had the same SecA binding capacity as MC1000 and SecE++ membranes (data not shown).Figure 3The effect of excess SecA on the kinetics of the translocation of OmpA across SecE-depleted membranes. A, the experiments were carried out as described under “Experimental Procedures,” except that purified SecA was added as indicated during translocation. B, translocations were carried out as described in A, with 20 μg/ml purified SecA at times as indicated. The reaction was stopped by addition of equal volume of 20% trichloroacetic acid, and the pellet was washed with ice-cold acetone once, dried by air, dissolved in sample buffer, and analyzed by SDS-PAGE. Data are mature form from the average of three independent experiments and presented as mean ± S.E., some of which not shown where variations are small. The translocation activity of MC1000 membranes at 10 min was used as 100%.View Large Image Figure ViewerDownload (PPT) The kinetics of translocation in SecE-depleted, SecE++, and MC1000 membranes were compared (Fig. 3 B). The translocation kinetic curves were virtually linear up to 5 min, indicating that SecE-depleted membranes are indeed active, albeit less efficient, in protein translocation. The ratio of translocation activity of SecE-depleted membranes reached 68% relative to MC1000 membranes after 10 min. To determine whether the apparent high activity of SecE-depleted membranes was due to the limited amount of nascent precursors, and since the kinetic curves in Fig. 3 B suggested that in the later time points this may be significant, saturation experiments were carried out. Membrane amounts used for the translocation assays were decreased from 25 μg to 5 μg, and titrated with post-translational35S-proOmpA supernatant (Fig. 4) to determine the saturation point. Translocation activities of MC1000 and SecE++ membranes showed that saturation was achieved with 140 μl of supernatant, and that of SecE-depleted membranes was 70 μl of supernatant. The translocation activity of SecE-depleted membranes with 140 μl of supernatant was 45% of wild-type membranes, this activity is comparable to that observed in typical translocation assays in which 5-fold more (25 μg) membrane was used (Fig.3 B). To further compare the translocation capacity of these membranes, competition experiments were carried out. As shown in Fig.5 A, the amounts of nonlabeled purified proOmpA needed for 50% competition in MC1000 and SecE-depleted membranes were 127 and 54 μg/ml, respectively. Thus, SecE-depleted membranes had about 40% capacity of wild-type membranes for proOmpA translocation, which is consistent with the translocation activity ratio mentioned above. The concentration of purified proOmpA used in this experiment was extremely high relative to nascent35S-proOmpA, which was estimated as 0.1 μg/ml in the posttranslational supernatant (data not shown). Since proOmpA was purified in 6 m urea, it may behave differently or may lose its competence for translocation when it is diluted 50-fold in translocation reaction. Therefore, similar experiments were carried out with nonlabeled nascent proOmpA. The competition results are similar to those with purified proOmpA (Fig. 5 B). Without additional SecA, SecE-depleted membranes had the same translocation activity as MC1000 membranes, and just slightly lower than SecE++ membranes with normal S30 (Fig. 2). Thus the competition experiment was also performed with nonlabeled purified proOmpA in the absence of SecA. As expected, there was no significant difference between competition curves of SecE++ and SecE− membranes (Fig. 5 C), supporting the previous observation that SecE-depleted membranes are active in translocation and the notion that the translocation characteristics of these membranes are similar. SecE-depleted membranes had reduced levels of other Sec proteins, except for SecA, which was increased. These membranes were still active in translocation of proOmpA without exogenous SecA (Fig. 2, lane 3), indicating that the SecA present in the membranes is sufficient for activities. To further examine the dependence of SecA in SecE-depleted membranes, the SecA inhibitor azide was added during translocation. The translocation activity of SecE-depleted membranes was significantly inhibited by azide (Fig. 6), indicating that translocation in SecE-depleted membranes remained SecA-dependent. In fact, the translocation of proOmpA in SecE-depleted membranes were more sensitive to azide than those in SecE++ and MC1000 membranes. In the presence of 2 μg of SecA, there was a 2-fold difference in I50 (50% inhibition) between these two membranes: the I50 was 1.7 mm in SecE-depleted membranes, and 3.7 mm in SecE++ membranes. The difference was even more pronounced in the presence of 6 μg of SecA; the I50 values of SecE-depleted and SecE++ membranes were 2.3 and 8 mm, respectively (data not shown). Translocation of proOmpA and proLamB are known to be SecB-dependent (36Kumamoto C.A. Beckwith J. J. Bacteriol. 1985; 163: 267-274Google Scholar, 37Kumamoto C.A. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 5320-5324Google Scholar, 38Kumamoto C.A. Francetic O. J. Bacteriol. 1993; 175: 2184-2188Google Scholar). In our in vitro system, the addition of purified SecB still enhanced the translocation of proOmpA and proLamB across both SecE++ and SecE− membranes (data not shown). This evidence also indicates that translocation across SecE− membranes is dependent on Sec system. SecA has been found to have cytoplasmic, peripheral, and integral membrane forms (39Cabelli R.J. Dolan K.M. Qian L. Oliver D.B. J. Biol. Chem. 1991; 266: 24420-24427Google Scholar). The integral form has been suggested to be the active form for translocation (4Watanabe M. Blobel G. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 9011-9015Google Scholar, 9Kim Y.J. Rajapandi T. Oliver D. Cell. 1994; 78: 845-853Google Scholar, 10Chen X. Xu H. Tai P.C. J. Biol. Chem. 1996; 271: 29698-29706Google Scholar). To determine what form of SecA is prevalent in SecE-depleted membranes, different membrane extractions were performed. Table II shows that about two-thirds of the membrane-bound SecA in SecE-depleted membranes were resistant to extraction by 0.1 m sodium carbonate (pH 11), 6 m urea, or 1 m heparin, suggesting that the two-thirds of SecA may be in the integral membrane form that is active for protein translocation.Table IISecA amounts in membranes treated with carbonate, heparin, and ureaTreatmentsSecA in membranesSecE++SecE−DTK buffer100100Na2CO362 ± 678 ± 8Heparin42 ± 169 ± 9Urea38 ± 775 ± 12Membranes were treated as described under “Experimental Procedures,” and resuspended in the same volume of DTK buffer. 10 μg of membrane was subjected to immunoblot for SecA quantitation. Data are presented as mean ± S.E. of three experiments for carbonate and two for urea and heparin treatments. SecA amounts of control membranes, the DTK buffer treated, were referred to 100%. Open table in a new tab Membranes were treated as described under “Experimental Procedures,” and resuspended in the same volume of DTK buffer. 10 μg of membrane was subjected to immunoblot for SecA quantitation. Data are presented as mean ± S.E. of three experiments for carbonate and two for urea and heparin treatments. SecA amounts of control membranes, the DTK buffer treated, were referred to 100%. To determine whether the substantial translocation activity in SecE-depleted membranes is limited to proOmpA, the effects of SecE depletion on the translocation of proLamB, proLpp, and proPhoA were examined in our in vitro system. It is notable that SecE++ membranes had higher activity than wild-type membranes for those four preproteins tested (Figs.3 B and 7). The activity ratios of SecE-depleted membranes to wild-type membranes for LamB and Lpp precursors were 50 and 38%, respectively, after 2 min of translocation reaction (Fig.7). After 10 min, the ratios changed to 80 and 63%, respectively. Thus, with a 50-fold decrease of SecE, the SecE-depleted membranes still contained substantial translocation activity for LamB and Lpp precursors, similar to proOmpA. Since OmpA, LamB, and Lpp all are outer membrane proteins, the kinetics of translocation of the soluble periplasmic protein, PhoA, was also examined. Interestingly, the translocation of proPhoA with SecE-depleted membranes was not as active as those of the other three precursor proteins. The activity of SecE-depleted membranes for proPhoA was only 23% of wild-type membrane after translocation for 2 min, and 36% after 10 min (Fig. 7). Thus the requirement of SecE or other Sec proteins in translocation may be precursor-dependent. It is unclear whether the destination of secretory protein is critical for SecE dependence in translocation. In SecE-depleted membranes, SecA increases 16-fold. The increased expression of SecA by depletion of SecE was predictable since thesecE gene was identified by SecA derepression (27Riggs P.D. Derman A.I. Beckwith J. Genetics. 1988; 118: 571-579Google Scholar), a common phenotype observed in most mutants with protein secretion defects (28Oliver D.B. Beckwith J. Cell. 1982; 30: 311-319Google Scholar,40Gardel C. Benson S. Hunt J. Michaelis S. Beckwith J. J. Bacteriol. 1987; 169: 1286-1290Google Scholar, 41Rollo E.E. Oliver D.B. J. Bacteriol. 1988; 170: 3281-3282Google Scholar). Genetic evidence has shown that SecE is functionally in excess and is the first membrane protein to interact with SecA (42Bieker K.L. Silhavy T. Cell. 1990; 61: 833-842Google Scholar, 43Bieker-Brady K. Silhavy T.J. EMBO J. 1992; 11: 3165-3174Google Scholar). The SecYEG complex has been proposed to be a receptor for SecA, existing as a high affinity binding site (6Hartl F. Lecker S. Schiebel E. Hendrick J.P. Wickner W. Cell. 1990; 63: 269-279Google Scholar, 7Economou A. Wickner W. Cell. 1994; 78: 835-843Google Scholar, 18Douville K. Price A. Eichler J. Economou A. Wickner W. J. Biol. Chem. 1995; 270: 20106-20111Google Scholar). However, our SecE-depleted membranes only contain 2% SecE, 15% SecY, and 17% SecG, but have 16 times more SecA. They also have the same binding capacity for exogenous SecA as wild-type membranes. Additionally, these SecA proteins are active in protein translocation (Fig. 2, lane 3), and most of them are found in integral form in membrane (Table II). Although the nature of the binding of SecA to SecE-depleted membranes is not certain, it is clear that the binding of SecA to the SecYEG complex is not obligatorily required for protein translocation. Since there is little SecE and much reduced SecYG in SecE-depleted membranes, it is also evident that the YEG membrane complex is not the essential membrane-embedded domain of preprotein translocase, in contrast to what has been proposed previously (16Brundage L. Fimmel C.J. Mizushima S. Wickner W. J. Biol. Chem. 1992; 267: 4166-4170Google Scholar, 18Douville K. Price A. Eichler J. Economou A. Wickner W. J. Biol. Chem. 1995; 270: 20106-20111Google Scholar). The interdependence of the Sec system is well documented. SecY overproduction is SecE-dependent (44Matsuyama S. Akimaru J. Mizushima S. FEBS Lett. 1990; 269: 96-100Google Scholar). However, SecE overproduction does not require the overproduction of SecY (45Tokuda H. Akimaru J. Matsuyama S. Nishiyama K. Mizushima S. FEBS Lett. 1991; 279: 233-236Google Scholar). This is in agreement with our data that SecE++ membranes have 12-fold more SecE but a normal amount of SecY. Additionally, SecF can stabilize SecD and SecY in an overproducing strain (46Sagara K. Matsuyama S. Mizushima S. J. Bacteriol. 1994; 176: 4111-4116Google Scholar). (On the other hand, depletion of SecD and SecF does not significantly influence the amounts of SecY and SecE (13Arkowitz R.A. Wickner W. EMBO J. 1994; 13: 954-963Google Scholar).) Furthermore, the amounts of SecY and SecG are also decreased in the SecE reduced background,secE501 (17Nishiyama K. Mizushima S. Tokuda H. EMBO J. 1993; 12: 3409-3415Google Scholar). Combining with our data that the depletion of SecE is accompanied by the depletion of other Sec proteins except for SecA, these results suggest an interdependence of all the Sec proteins. Traxler and Murphy (30Traxler B. Murphy C. J. Biol. Chem. 1996; 271: 12394-12400Google Scholar) have shown that integral membrane proteins are not inserted and degraded even under mild SecE-depleted condition, providing a possible explanation for the reduction of other Sec proteins. SecE has been reported to be essential for protein translocation across reconstituted proteoliposomes providing the basis for current dogma that SecYEG forms the translocation channel (15Brundage L. Hendrick J.P. Schiebel E. Driessen A.J.M. Wickner W. Cell. 1990; 62: 649-657Google Scholar, 21Akimaru J. Matsuyama S. Tokuda H. Mizushima S. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 6545-6549Google Scholar, 45Tokuda H. Akimaru J. Matsuyama S. Nishiyama K. Mizushima S. FEBS Lett. 1991; 279: 233-236Google Scholar). It is not yet clear why the requirements of SecE in vitro are different in various protein translocation systems. The translocation of SecE-depleted membranes is at least as active as these reconstituted systems. However, the efficient processing of secretory precursors with these reconstituted proteoliposomes has never been reported. Furthermore, the possibility of functional substitution of any Sec protein by another protein has not been ruled out. It is possible that there is another protein(s) that can replace the function of SecE; some other non-Sec protein(s) may be involved in active translocation across SecE-depleted membranes. First, YajC, a product encoded from the gene located withinsecD operon, has been shown to interact with SecY (47Taura T. Akiyama Y. Ito K. Mol. Gen. Genet. 1994; 243: 261-269Google Scholar). Second, phage shock protein A (PspA), which is induced in response to heat, ethanol, osmotic shock, and infection by filamentous bacteriophages (48Brissette J.L. Russel M. Weiner L. Model P. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 862-866Google Scholar), has been reported to have a stimulatory effect on protein export in E. coli (49Kleerebezem M. Tommassen J. Mol. Microbiol. 1993; 7: 947-956Google Scholar). Third, another new gene product, Syd (suppressor of secY dominance), has been found to suppress secY −d1 mutant, and to stabilize overproduced SecY (50Shimoike T. Taura T. Kihara A. Yoshihisa T. Akiyama Y. Cannon K. Ito K. J. Biol. Chem. 1995; 270: 5519-5526Google Scholar). The possible involvement of these proteins in translocation across SecE-depleted membranes needs to be investigated. Translocation channels in SecE-depleted membranes seem to have different efficiencies for translocating different proteins. The translocation activities of proOmpA, proLamB, and proLpp are only moderately inhibited in SecE-depleted membranes even in the presence of excess SecA. However, for precursor of PhoA (alkaline phosphatase), the translocation activity is reduced to 20–36% relative to wild-type membranes, implying that SecE, or other Sec proteins, might be required for the translocation of proPhoA. Alternatively, some other factor(s) needed for translocation of proPhoA might be reduced in SecE-depleted membranes. Indeed, protein export was found to be severely defectivein vivo with the same strain used here, PS289 (30Traxler B. Murphy C. J. Biol. Chem. 1996; 271: 12394-12400Google Scholar). The cold-sensitive SecE mutants raise an interesting question. ThesecE mutation is lethal to the cell at nonpermissive temperatures. Some mutants express less than 50% SecE even at permissive temperatures (25Schatz P.J. Bieker K.L. Ottemann K.M. Silhavy T.J. Beckwith J. EMBO J. 1991; 10: 1749-1757Google Scholar). However, cold sensitivity has been found to be an inherent property of the secretion pathway itself (51Pogliano K.J. Beckwith J. Genetics. 1993; 133: 763-773Google Scholar). The reduced synthesis of SecE in these mutants, therefore, may not be the direct reason for lethality at nonpermissive temperatures, since we also found that the translocation activity of SecE-depleted membranes is not cold-sensitive (data not shown). (It has been reported more recently that the secE gene is absent in the genome ofMycoplasma genitalium (52Fraser C.M. Gocayne J.D. White O. Adams M.D. Clayton R.A. Fleischmann R.D. Bult C.J. Kerlavage A.R. Sutton G. Kelley J.M. Fritchman J.L. Weidman J.F. Small K.V. Sandusky M. Fuhrmann J. Nguyen D. Utterback T.R. Saudek D.M. Phillips C.A. Merrick J.M. Tomb J. Dougherty B.A. Bott K.F. Hu P. Lucier T.S. Peterson S.N. Smith H.O. Hutchison III, C.A. Venter J.C. Science. 1995; 270: 397-403Google Scholar) and Methanococcus jannaschii (53Bult C.J. White O. Olsen G.J. Zhou L. Fleischmann R.D. Sutton G.G. Blake J.A. FitzGerald L.M. Clayton R.A. Gocayne J.D. Kerlavage A.R. Dougherty B.A. Tomb J. Adams M.D. Reich C.I. Overbeek R. Kirkness E.F. Weinstock K.G. Merrick J.M. Glodek A. Scott J.L. Geoghagen N.S.M. Weidman J.F. Fuhrmann J.L. Nguyen D. Utterback T.R. Kelly J.M. Peterson J.D. Sadow P.W. Hanna M.C. Cotton M.D. Roberts K.M. Hurst M.A. Kaine B.P. Borodovsky M. Klenk H. Fraser C.M. Smith H.O. Woese C.R. Venter J.C. Science. 1996; 273: 1058-1073Google Scholar), both of which are among the smallest free-living organisms. However, this needs to be further confirmed.) If SecE is not essential for translocation, why does cell growth halt when SecE is depleted? One possibility is that SecE may have another physiological function(s) relative to cell growth. In other words, SecE depletion might cause secondary effects, which lead to cessation of growth. Thus, it may be possible that some other proteins, which are directly involved in cell growth, are affected by depletion of SecE. Alternatively, we favor the notion that, like PhoA shown here, translocation of some other essential proteins are more strictly dependent on SecE. In this case, the specific dependence can explain the cessation of cell growth in vivo after SecE depletion, and the genetic evidences that SecE is essential. Even with a reduction of SecE down to 2%, SecE-depleted membranes are still active in translocation of proOmpA, proLamB, and proLpp, at about 50% of the activity of wild-type membranes in the presence of excess SecA, and at an even higher ratio in its absence. Therefore we conclude that SecE is not absolutely essential for in vitro translocation of, at least, certain secretory preproteins (but probably not all precursors). Then, which Sec protein is responsible for the translocation activity in SecE-depleted membranes for these precursors? SecD and SecF have been shown to only have an enhancing effect on translocation (14Matsuyama S. Fujita Y. Sagara K. Mizushima S. Biochim. Biophys. Acta. 1992; 1122: 77-84Google Scholar), and so does SecG (20Nishiyama K. Hanada M. Tokuda H. EMBO J. 1994; 13: 3272-3277Google Scholar). Even the requirement for SecY, which has 10 transmembrane segments, is still controversial, since it has been shown that translocation can occur with reconstituted membranes depleted of SecY (4Watanabe M. Blobel G. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 9011-9015Google Scholar, 23Watanabe M. Nicchitta C.V. Blobel G. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 1960-1964Google Scholar), and we have confirmed this as well. 2Y.-B. Yang, J. P. Lian, and P. C. Tai, manuscript submitted for publication. Thus, the best candidate for translocation activities in SecE-depleted membranes is SecA. We have recently found that some SecA is permanently embedded in membranes in an active integral form and may form part of the translocation channel (10Chen X. Xu H. Tai P.C. J. Biol. Chem. 1996; 271: 29698-29706Google Scholar). Therefore, we propose that SecA is the only indispensable component of the Sec machinery, although it may not be sufficient alone. It is possible that, together with all other Sec proteins in proper ratios as in normal membranes, SecA can translocate proteins more efficiently. However, when other Sec proteins are diminished, SecA is still active in transporting proteins, albeit with less efficiency. This interpretation is analogous to earlier observations on the energy requirement for protein translocation; ATP is essential, but less efficient, as energy source for translocation without proton motive force (54Chen L.L. Tai P.C. Proc. Natl. Acad. Sci. U. S. A. 1985; 82: 4384-4388Google Scholar, 55Chen L.L. Tai P.C. J. Bacteriol. 1986; 167: 389-392Google Scholar, 56Geller B. Movva N.R. Wickner W.P. Proc. Natl. Acad. Sci. U. S. A. 1986; 83: 4219-4222Google Scholar). Interestingly, ATP is required for SecA function. Thus most of the translocation activity of SecE-depleted membranes may come from the 16-fold increase in SecA, in conjunction with other reduced Sec proteins, as active translocation sites with reduced efficiency. That may explain why SecE-depleted membranes are more sensitive to azide than SecE++ and MC1000 membranes, and why their translocation activity cannot be enhanced as much as wild-type membranes by exogenous SecA, even though both membranes have the same binding capacity for it. Thus, while SecE depletion may have caused severe defects leading to cessation of cell growth, it is also clear from the data presented here that SecE is not essential for translocation of some proteins but may be required for other precursors. It follows that since the YEG complex is not obligatory for protein translocation in general, the active integral form of SecA may play a more important role than previously realized in forming the translocation channel in the inner membrane. We thank J. Beckwith, P. Schatz, D. Oliver, and C. Murphy for strains, plasmids, and SecE peptide antibodies; P. Li for establishing the in vitro cotranscription-translation system; and A. Boyer for reading of the manuscript. We also acknowledge the numerous discussions with the laboratory members who also contributed purified proteins, plasmids, and antibodies." @default.
• W2041316200 created "2016-06-24" @default.
• W2041316200 creator A5012396051 @default.
• W2041316200 creator A5027438427 @default.
• W2041316200 creator A5037231508 @default.
• W2041316200 date "1997-05-01" @default.
• W2041316200 modified "2023-10-12" @default.
• W2041316200 title "SecE-depleted Membranes of Escherichia coli Are Active" @default.
• W2041316200 cites W145694369 @default.
• W2041316200 cites W1481018570 @default.
• W2041316200 cites W1504676920 @default.
• W2041316200 cites W1511816051 @default.
• W2041316200 cites W1529595022 @default.
• W2041316200 cites W1584878924 @default.
• W2041316200 cites W1590719173 @default.
• W2041316200 cites W1610704689 @default.
• W2041316200 cites W1626287517 @default.
• W2041316200 cites W1644489643 @default.
• W2041316200 cites W17566691 @default.
• W2041316200 cites W1848937193 @default.
• W2041316200 cites W1923570112 @default.
• W2041316200 cites W1932645010 @default.
• W2041316200 cites W1943148606 @default.
• W2041316200 cites W1946720111 @default.
• W2041316200 cites W1964517070 @default.
• W2041316200 cites W1965731203 @default.
• W2041316200 cites W1971302592 @default.
• W2041316200 cites W1973764696 @default.
• W2041316200 cites W1975085160 @default.
• W2041316200 cites W1992470956 @default.
• W2041316200 cites W2005038672 @default.
• W2041316200 cites W2005740894 @default.
• W2041316200 cites W2006029448 @default.
• W2041316200 cites W2008223866 @default.
• W2041316200 cites W2022217156 @default.
• W2041316200 cites W2023506192 @default.
• W2041316200 cites W2035366397 @default.
• W2041316200 cites W2037341533 @default.
• W2041316200 cites W2041652393 @default.
• W2041316200 cites W2049434532 @default.
• W2041316200 cites W2053211842 @default.
• W2041316200 cites W2055229289 @default.
• W2041316200 cites W2059697996 @default.
• W2041316200 cites W2066537459 @default.
• W2041316200 cites W2067410749 @default.
• W2041316200 cites W2073075045 @default.
• W2041316200 cites W2073492171 @default.
• W2041316200 cites W2073725165 @default.
• W2041316200 cites W2080304004 @default.
• W2041316200 cites W2084231685 @default.
• W2041316200 cites W2084334026 @default.
• W2041316200 cites W2095153757 @default.
• W2041316200 cites W2096207342 @default.
• W2041316200 cites W2097932259 @default.
• W2041316200 cites W2100837269 @default.
• W2041316200 cites W2126106174 @default.
• W2041316200 cites W2150771080 @default.
• W2041316200 cites W2152737402 @default.
• W2041316200 cites W2203768638 @default.
• W2041316200 cites W2338801551 @default.
• W2041316200 cites W247603372 @default.
• W2041316200 cites W40140278 @default.
• W2041316200 cites W4239833500 @default.
• W2041316200 cites W4293247451 @default.
• W2041316200 doi "https://doi.org/10.1074/jbc.272.21.13660" @default.
• W2041316200 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/9153216" @default.
• W2041316200 hasPublicationYear "1997" @default.
• W2041316200 type Work @default.
• W2041316200 sameAs 2041316200 @default.
• W2041316200 citedByCount "38" @default.
• W2041316200 countsByYear W20413162002012 @default.
• W2041316200 countsByYear W20413162002013 @default.
• W2041316200 countsByYear W20413162002015 @default.
• W2041316200 countsByYear W20413162002017 @default.
• W2041316200 countsByYear W20413162002018 @default.
• W2041316200 countsByYear W20413162002019 @default.
• W2041316200 countsByYear W20413162002020 @default.
• W2041316200 countsByYear W20413162002021 @default.
• W2041316200 countsByYear W20413162002023 @default.
• W2041316200 crossrefType "journal-article" @default.
• W2041316200 hasAuthorship W2041316200A5012396051 @default.
• W2041316200 hasAuthorship W2041316200A5027438427 @default.
• W2041316200 hasAuthorship W2041316200A5037231508 @default.
• W2041316200 hasBestOaLocation W20413162001 @default.
• W2041316200 hasConcept C104317684 @default.
• W2041316200 hasConcept C12554922 @default.
• W2041316200 hasConcept C185592680 @default.
• W2041316200 hasConcept C41625074 @default.
• W2041316200 hasConcept C547475151 @default.
• W2041316200 hasConcept C55493867 @default.
• W2041316200 hasConcept C86803240 @default.
• W2041316200 hasConceptScore W2041316200C104317684 @default.
• W2041316200 hasConceptScore W2041316200C12554922 @default.
• W2041316200 hasConceptScore W2041316200C185592680 @default.
• W2041316200 hasConceptScore W2041316200C41625074 @default.
• W2041316200 hasConceptScore W2041316200C547475151 @default.
• W2041316200 hasConceptScore W2041316200C55493867 @default.
• W2041316200 hasConceptScore W2041316200C86803240 @default.

• next